ML20045F409
| ML20045F409 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 07/02/1993 |
| From: | Feigenbaum T NORTH ATLANTIC ENERGY SERVICE CORP. (NAESCO) |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| NYN-93098, TAC-M85020, NUDOCS 9307070279 | |
| Download: ML20045F409 (10) | |
Text
..
q 8
4 Odh fee ~btx5?"03e74 Telephone (603)474-9521
=
h
(.
Facsimile (603)474-2987-Energy Service Corporation Ted C. Feigenbaum Senior Vice President and Chief Nuclear Officer NYN-.93098 July 2,1993 United States Nuclear Regulatory Commission Washington, D.C. 20555 Attention:
Document Control Desk
References:
(a)
Facility Operating License No. NPF-86, Docket No. 50-443 (b)
NRC Letter, G. Edison to Docket No. 50-443 dated February 27,1992,
" Summary of Meeting With Public Service Company of New Hampshire Regarding Replacement of the Movable Incore Detector System with a Fixed Incore Detector System at Seabrook Statio (TAC NO. M82546)"
t (c)
North Atlantic Letter NYN-92162, dated November 25,1992. " License Amendment Request 92-14: Incore Detector System", T. C. Feigenbaum to USNRC (d)
USNRC Letter dated May 28, 1993, " Request for Additional Information (TAC Nm MM85020)",
A.
W. De Agazio to T.
C.
l' Feigenb - :,
Subject:
Response to Requ.
for Additional Information:
License Amendment Request 92-14 (TAC No. M85020)
Gentlemen:
North Atlantic Energy Service Corporation (North Atlantic) provides in the Enclosure, the response to the Request for Additional Information forwarded by Reference (d). North Atlantic believes that this information will support'the completion of the review of License Amendment Request 92-14 [ Reference (c)]. Should additional questions arise.
concerning the YAEC Reports, North Atlantic will be available to meet with the Staff to provide the required explanation.
Should you have any questions regarding this letter, please contact Mr. Terry L.
Harpster, Director of Licensing Services, at (603) 474-9521, extension 2765.
Very truly yours,
(
w Ted Feigenbaum.
TCF:JMP/ tad Enclosure
(;60141 y
- I
[f a member of the Northeast Utilities system
-i l;
9307070279 930702 PDR ADOCK 05000443 j
OQ PDR y.1
.~
~,
h
.y l
l-:
1
~
United States Nuclear Regulatory Commission July 2,' 1993
.j
. Attention:
Document Control Desk Page two cc:
Mr. Thomas T. Martin
-f Regional Administrator United States Nuclear Regulatory Commission Region I 475 Allendale Road King of Prussia, PA 19406 Mr. Albert W. De Agazio, Sr. Project Manager Project Directorate I-4 Division of Reactor Projects U.S. Nuclear Regulatory Commission Washington, DC 20555 Mr. Noel Dudley
[
NRC Senior Resident Inspector P.O. Box 1149 Seabrook, NH 03874 i
Mr. George L. Iverson, Director New Hampshire Office of Emergency Management j
107 Pleasant Street Concord, NH 03301 I
s ur
)
6 b
E
+
)
3 8
r p
1
~~+ev-.
,,4 s
North Atlantic July 2,1993 I
i ENCLOSURE TO NYN-93098 RESPONSE TO REOUEST FOR ADDITIONAL INFORMATION (TAC NO. M85020) i i
s i
E
1 l
l l
I REOUESTS FOR ADDITIONAL INFORM ATION
.)
i 1
I 1.
Describe your software QA testing program for the fixed incore detector system
]
(FIDS) data acquisition and processing software.
RESPONSE
l The Fixed Incore Detector System at Seabrook Station utilizes two software systems to process and analyze the power distribution in the reactor. The first software system, designed by North Atlantic, is for data acquisition of the 290 fixed detector signals and is called the Fixed Incore Detector Data Acquisition System (FIDDAS). The second software system, designed by Yankee Atomic Electric Company (YAEC), is a series of codes that j
analyze the fixed incore detector signals and develop the reactor power distribution. All of 1
this software was developed and tested under the appropriate software development and quality assurance procedures of North Atlantic and YAEC.
The purpose of the FIDDAS is to collect and store the 290 fixed incore detector signals once per minute. In addition, the system collects and stores a number of reactor operating parameters (e.g., power, rod position, axial flux difference) each time the fixed incore detector signals are read. The FIDDAS operates as part of the Main Plant Process Computer System. Both the hardware and software for the FIDDAS were designed, installed and tested prior to initial plant startup and have been operating since the plant commenced operation. The FIDDAS was subjected to a number of tests to ensure that it was performing its intended function. Testing included verification of detector voltage readings through the 1
use of simulated detector signals and verification of proper detector addressing, data transfer and data storage. The testing is documented in the FIDDAS Acceptance Test Plan and Startup Test Procedure GT-I-103, Fixed Incore Detector System.
The second software system is called the Reactor Analysis Workstation (RAWy. The purpose of the RAW is to process the fixed incore detector data collected by the FIDDAS and infer the reactor power distribution and surveillance parameters. The RAW consists of three main segments; SIMULATE-3, the Fixed Incore Detector Analysis Code (FINC) and a set of linking and processing codes. The development of the RAW software was performed by i
YAEC in accordance with their approved Yankee Operational Quality Assurance Program as implemented by the procedures in the Yankee Engineering Manual. The Engineering Manual procedures require the development of the software functional requirements, a description of the methodology employed, preparation of user and programmer manuals and
)
validation of proper code operation. All of the above features receive independent review j
and approval prior to issuance of the code for use.
SIMULATE-3 is used to determine the predicted fixed incore detector signals and incore power distribution. These predicted signals are then used as input to the FINC code.
SIMULATE-3 has been reviewed and approved for use in reload licensing analysis by the NRC. Software modifications to SIMULATE-3 required for the modeling of the platinum detectors is discussed in the response to Question 2. Acceptance testing of SIMULATE-3 by YAEC was performed by obtaining acceptable results to a battery of vendor supplied test j
cases, as well as specific test cases developed by YAEC.
The FINC code uses the predicted fixed incore detector signals developed by SIMULATE-3 and the actual fixed incore detector signals collected by the FIDDAS to infer the reactor power distribution and local peaking factors.
The methodology used in this code is described in topical report YAEC-1855P, Seabrook Station Unit 1 Fixed Incore Dectector System (FIDS) Analysis, currently under NRC review. The FINC code was developed and 1
I l'
=
1 i
l 1
documented according to the Yankee Operational Quality Assurance Program and Yankee l
Engineering Manual procedures described above. A series of test cases to demonstrate the proper operation of the code are defined and anticipated results are documented. These
)
tests were executed on the RAW and the results were verified.
i The third segment of the RAW is a series of three linking and processing codes that manage input and output files for SIMULATE-3 and FINC. The first code searches the background data collected by FIDDAS to determine how many times steps are required to keep the SIMULATE-3 model current with respect to core operating history. The second code creates the SIMULATE-3 input files for each of the time steps determined above. The final code summarizes key SIMULATE-3 results for review.
The functional requirements, methodology, validation, user and programmer manuals are documented according to the Yankee Operational Quality Assurance Program procedures.
}
The final steps in verification of the fixed incore detector system software was an integral test of the software on the RAW. The purpose of this test was to verify proper installation and operation of the fixed incore detector software systern. This test was conducted satisfactorily and also documented according to the Yankee Operational Quality Assurance Program procedures.
i The results of the individual software quality assurance tests and other referenced material are available for review at Seabrook Station.
2.
Describe the software modifications required for modeling Platinum (Pt) detectors that were made to the previously reviewed CASMO-3/ SIMULATE-3 code package and discuss any benchmarking results.
[
t it ES PONS E:
No modifications were made to the CASMO-3, infinite lattice cross-section code. The code accepts, as input, gamma detector response values as a function of gamma flux energy group.
This method is used for all gamma detector response calculations within CASMO-3, regardless of the detector material. The default gamma response function within CASMO-3 is for iron. The iron response was approved for use with BWR gamma sensitive detectors.
SIMULATE-3 was modified, by Studsvik of America, to allow the user to input the fractional neutron component of the detector signal. This fraction is given in terms of the total detector signal, and it can be distributed by either or both the fast and thermal neutron flux.
The gamma portion of the detector's signal is determined through the total responses determined in CASMO-3 cases and local detector neutron flux calculations within.
SIMULATE-3.
This is the standard method of detector calculations used within SIMULATE-3. The total neutron portion of the detector signalis determined from the input fraction and is then distributed by the SIMULATE 3 calculated relative local neutron flux levels at the detector locations. The individual detector's gamma and neutron portions are then summed to determine the detector's total signal.
1 No specific benchmarks are available to confirm the neutron contribution methodology. The assumption that 25% of the total detector signal is due to thermal neutrons was determined following a literature search and sensitivity study where this value was varied from 20% to l
30%. The results of this study are provided in YAEC-1855P, Seabrook Station Unit 1 Fixed incore Dectector System (FIDS) Analysis, (see response to Question 3).
The streaming of gamma rays on the periphery of the core is not modeled in the infinite lattice calculations in CASMO-3. SIMULATE-3 was modified, by Studsvik of America, to i
2
~l e
-~
n
t i
C i
compensate for' gamma leakage for peripheral assemblies. This correction for gamma source to gamma flux is based on a one-group solution to the gamma diffusion equation with a
- fourth-order source distribution in each node. The gamma source distribution is developed from the local neutron flux distribution.
j No specific benchmarks were used to verify the gamma periphery leakage correction. The gamma leakage correction was tested at YAEC by comparing cases with and without this
.I option turned on.
The results of the analysis showed that this correction reduces core l
average measured to predicted detector ratio, in all cases.
j The results of the analysis and other referenced material are available for review at.
[
Seabrook Station.
I 3.
You state that approximately 25% of the total detector signalis derived from neutron j
flux response. Ilow much of the signal is due to the response to thermal neutrons j
alone?
Ilow does the thermal and total neutron flux response vary with core residence lifetime? How are multibundle effects represented?
}
f
RESPONSE
The entire 25% of the total detector signal derived from the neutron flux is attributed to the i
thermal flux. No response due to the fast neutron flux is assumed. No information to the i
contrary was found during a literature search of the topic. The justification of the YAEC
[
assumption is evident in the results of the sensitivity study performed to determine the appropriate percentage of total signal to attribute to the neutron flux, as documented in j
YAEC-1855P, Seabrook Station Unit 1 Fixed Incore Dectector System (FIDS) Analysis.
The magnitude of the neutron response is derived from the core total gamma portion of the detector signal.
The gamma portion of the detector signal is calculated ' from the SIMULATE-3 local neutron flux and the exposure dependent CASMO-3 gamma detector l
response data. The core total neutron portion of the det.ector signal, will be distributed to
}
the detectors by the relative ~ thermal flux at the local detectors. The incore residence effects on detector neutron flux response are determined as a function of the local neutron flux.
~!
The local neutron flux is accurately calculated by the CASMO-3/ SIMULATE-3 methodology.
j I
The gamma leakage correction given in SIMULATE-3, (see Question 2), is the only multibundle effect represented for the gamma portion of the detector's signal.
The j
multibundle effects of the neutron flux are accounted for by using the SIMULATE-3 local i
detector flux, which takes into account all effects within the core. The ' relative neutron thermal flux in the local detector is used to determine the local neutron contribution to the
[
total detector signal.
t 4.
Is it necessary to treat the effects of Platinum depletion? If not, please present a quantitative numerical analysis on why this effect may be neglected.
RESPONSE
The modeling of the effects of Platinum (Pt) depletion is not required. The expected rate of depletion of the platinum detectors due to neutron absorption can be calculated from the j
Table of Isotopes. The reactions of interest are given below:
3
l j
Nat ural Thermal AbundanceL Resulting Neutron Cross Isotope
(%)
Reaction Product Section (barns) 4 Pt-190 0.01 (n, E.C.)
Ir-191 800.00 l
Pt-192 0.79 (n,y)
Pt-193*
10.00 Pt-194 32.90 (n,y)
Pt-195 1.20 j
f j.
Pt-195 33.80 (n,y)
Pt-196 29.00 Pt-196 25.30 (n,p)
Au-197 0.75 i
Pt-198 7.20 (n,p)
A u-199 3.80 Pt 193 has a 50 year half-life and so we will assume it is stable for the purposes of this calculation.
j The only reactions that result in non-Platinum nuclei are those for Pt-190, Pt-196, and Pt--
198. The total
- effective" c'ross-section for such reactions is then-f I
o = 0. 0 001 x8 0 0. 0 0 +0. 2 53 0 x0. 7 5 +0. 07 2 0 x3. 8 0 =0. 54 barns i
l l
The total number of absorptions per second that result in non-Pt nuclei is then c
i Absorptions,y,y,,,4 Second
)
L Atom density of Pt (atoms /cc)
)
l where:
N
=
Total volume of Pt (cc) j V
=
0.54X 10 (cm')
l o
=
Thermal neutron flux (n/cm'/sec)
=
The percent atoms that ' deplete" per second is then l
Absorptions Absorptions I
sec sec x100=
x100=ox$x100
=
sec Tota 1 Atoms NxV In a typical PWR environment, & is about SX10'8 n/cm'/sec. Using a = 0.54 barns, we have
=0. 54 x10-8'x5. 00 x10"x100 =2.7 0x10 =0. 08 5 sec year Thus, the rate of depletion will be less than 0.1% per year or less than 1% in ten years.
These results show that the loss of sensitivity due to depletion of the Platinum material will be insignificant.
l l
i 4
i I
1
5.
Explain why the ratios of fixed / movable inferred ' measured' Fxy (Table 5.6) and Fq (Table 5.7) decrease with cycle exposure. Does this indicate a burnup dependent bias between the fixed and movable incore detector signal to-power conversion methodologies?
l
RESPONSE
?
The differences between inferred power parameters from the fixed detector calculations and Westinghouse movable detector calculations are derived from the inherent differences in the reactor physics methods used to predict the power distribution.
t The decrease in ratio of fixed / movable power parameters is not isolated only to radial peaking factor (Fxy) and the heat flux hot channel factor (Fq). The inferred values of nuclear enthalpy rise hot channel factor (Fdh) also exhibit this trend, as can be seem in the top graphs provided in Figures 1 and 2.
The trends for Fdh should be similar to trends in other peak power parameters as the Fxy and Fq are usually in the same assembly locations.
The inferred Fdh value is determined from the ratio of measured to predicted detector signals and the prediction of Fdh from the reactor physics methodology. The measured to predicted detuctor signal ratio and prediction of Fdh are a function of the reactor physics methodology and detector system.
i t
The difference between the Westinghouse and YAEC reactor physics calculations is provided in the middle graphs of Figure 1 and 2. For Cycle 1, this plot shows that the value of Fdh predicted by Westinghouse reactor physics methods becomes greater than that predicted by YAEC reactor physics methods, as the cycle is depleted. The graph of Cycle 2, shows the same trend with less magnitude. Thus, the expected result is for the ratio,'of inferred Fdh.
from Westinghouse movable detector system to that measured by the fixed detector system, to decrease as the core is depleted, i
The ratio of measured to predicted values of Fdh, is provided for both Westinghouse and j
YAEC in the bottom graph of Figures 1 and 2. For Cycle 1, this data suggests that both the fixed and movable detector systems inferred Fdh values greater than predicted. As can be j
seen, there is little difference in this ratio for the two systems.
The Cycle 2 data suggests that when using Westinghouse methods, the movable system inferred Fdh is increasing, while the predicted Fdh is decreasing. However, the inferred Fdh from the movable detectors system processed with YAEC methods, show that the Fdh is decreasing as the core is depleted. This can be seen in the bottom graph of Figure 2. This discrepancy can therefore be attributed to the small differences in the two reactor physics method's to predict detector signals and local peaking factors.
Except for a few points, comparisons between Fdh, Fxy and Fq values inferred by the fixed h
and movable detector systems, which were both processed with predictions from YAEC methods (SIMULATE-3), show very good agreement. The major exception being that the l
fixed detector system consistently produces Fdh values greater than that of the movable syst em. Thus, the two detector systems yield similar results, when the measurements are processed with predictions from the same reactor physics methodology.
I l
l 5
i i
Figure 1 Seabrook Station Cycle 1 Ratio of Fixed Systern Value to Westinghouse Movable System Value 1.06 1.04 -
/
]
1.02 -
e O
b W
1.0 - ----------
m 6
0.98 -
0.96 -
o ra,o g usosa,,,e ren vanw nxee i wesungwse mvame
~
o na: e usasureo rry Yanw Fmed /Wesungwse MovnMe A Raf to of Measured Fq Yankee Raed / Wesurghouse Movabie 0.94 S
0 1
2 3
4 5
6 7
8 9
10 11 12 13 Ratio of Yankee Predicted Fdh to Westinghouse Predicted Fdh 1.04 -
1.02 -
o 18 1.0 - ----------
1 0.98 -
0.96 -
, nono e e,saaserenY.n fwe.1,ngnouse 0.94 0
1 2
3 4
5 6
7 8
9 10 11 12 13 Ratio of Westinghouse and Yankee Measured to Predicted Fdh 1.06 1.04 -
m f
1.02 -
}
7 f-
~
O 1.0 - -------------~~------------------------------------------------------------------
t i
0.98 -
t 0.96 -
- rwo ot r on wesungnouse uavaoie usasurso r wesiingnouse preocoa e Ramo d f dn YanW Tased Measured / Yankee Predded 0.94 0
1 2
3 4
5 6
7 8
9 10 11 12 13 t
Exposure (GWd'MTu)
Figure 2 Seabrook Station Cycle 2 Ratio of Fixed System Value to Westinghouse Movable Systern Value 1.04 -
C A
1.02 -
'T OR 1.0 - -------------------~~---------
~~---------------------
x
~
~
0.98 -
0.96 -
o rw.o or usasu ee ron Yanw nxed iwesungnouse uovame o rwe of masuree r,y Yankee rumo r wesungnouse mvam.
A rum of Measured rq Yankee rimediWestinghouse Movab:e 0.94 0
1 2
3 4
5 6
7 8
9 10 Ratio of Yankee Predicted Fdh to Westinghouse Predicted Fdh 1.06 1.04 -
1.02 -
o15 1.0 -----------------------------~~------------------- 7 t
0.98 -
0.96 -
, rwe of preoceo ran Yankee iwesongnouse 0.94 0
1 2
3 4
5 6
7 8
9 10 Ratio of Westinghouse and Yankee Measured to Predicted Fdh 1.06 1.04 -
N 3
1.02 -
\\
2..
\\-
=
O
% 1.0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ~ ~ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ~ ~ ~ ~ - - - - -
C 0.98 -
0.96 -
rwe of ret wounanoase mvame mauri,o iwanng+ maw preoceo e rwm of ron Yankee razed Maasured i Yankee Predced a rwe of rdh Yankee Movede masured i Yankee Prerncted 0.94 0
1 2
3 4
5 6
7 8
9 10 Exposure (GWd/MTu)